- 06 Jan, 2024 1 commit
-
-
Kirthi Shankar Sivamani authored
* Deterministic FA, bump minimum supported version Signed-off-by:
Kirthi Shankar Sivamani <ksivamani@nvidia.com> * fixes Signed-off-by:
-
- 05 Jan, 2024 1 commit
-
-
Przemyslaw Tredak authored
Signed-off-by:Przemek Tredak <ptredak@nvidia.com>
-
- 03 Jan, 2024 3 commits
-
-
Przemyslaw Tredak authored
Signed-off-by:
-
Sangkug Lym authored
* Provide pre-computed max sequence to remove unnecessary kernels and D2H copies Signed-off-by:
Sangkug Lym <slym@nvidia.com> * Tweak comments Signed-off-by:
Tim Moon <4406448+timmoon10@users.noreply.github.com> --------- Signed-off-by:
Tim Moon <tmoon@nvidia.com> Co-authored-by:
-
Przemyslaw Tredak authored
Signed-off-by:
-
- 02 Jan, 2024 1 commit
-
-
Hongbin Liu authored
avoid redundant computation for cu_seqlens Signed-off-by:
Hongbin Liu <hongbinl@nvidia.com> Co-authored-by:
-
- 18 Dec, 2023 1 commit
-
-
Alp Dener authored
* Linear and LayerNormLinear weight and bias buffer cleanup at the end of init when there is no parameter split Signed-off-by:
Alp Dener <adener@nvidia.com> * fixed typo in tensor name Signed-off-by:
-
- 16 Dec, 2023 2 commits
-
-
cyanguwa authored
* add sliding window to FA Signed-off-by:
Charlene Yang <8636796+cyanguwa@users.noreply.github.com> * fix forward logic Signed-off-by:
-
Tim Moon authored
* Update fp8_meta amax when copying into Float8Tensor Signed-off-by:
-
- 15 Dec, 2023 3 commits
-
-
Przemyslaw Tredak authored
* Disable dynamo for Fused Attention Signed-off-by:
-
Kunlun Li authored
Avoid redeclaration error of "nv_bfloat16" when Compiling in CUDA12.1. Error log: /usr/local/cuda/include/cuda_fp16.hpp(2724): error: invalid redeclaration of type name "nv_bfloat16" (declared at line 2837 of /usr/local/cuda/include/cuda_bf16.hpp) Signed-off-by:Kunlun Li <94586211+kunlunl@users.noreply.github.com>
-
Fabian Joswig authored
[fix] fixed micro batched inference with RoPE Signed-off-by:
Fabian Joswig <fabian.joswig@deepl.com> Co-authored-by:
-
- 13 Dec, 2023 1 commit
-
-
Marks101 authored
Signed-off-by:Markus Schnoes <markus.schnoes@gmx.de>
-
- 12 Dec, 2023 2 commits
-
-
cyanguwa authored
disable pylint bare except Signed-off-by: -
cyanguwa authored
* fix onnx/dynamo error Signed-off-by:
Przemyslaw Tredak <ptrendx@gmail.com> Signed-off-by:
-
- 07 Dec, 2023 2 commits
-
-
cyanguwa authored
* Integrate cuDNN frontend v1 to fused attention and miscellaneous fixes Signed-off-by:
-
Tim Moon authored
* Float8Tensor uses cached transpose if available Signed-off-by:
-
- 04 Dec, 2023 1 commit
-
-
Marks101 authored
* [PyTorch] TransformerLayer: add parallel_attention_mlp to support Falcon models Signed-off-by:
-
- 01 Dec, 2023 2 commits
-
-
Tim Moon authored
Fix incorrect variable name in LayerNormMLP backward Signed-off-by: -
LadyRick authored
[PyTorch] fix amax calculate during fp8 calibration Signed-off-by:ladyrick <ladyrick@qq.com>
-
- 30 Nov, 2023 1 commit
-
-
Deepak Narayanan authored
wgrad should be zero'ed out if a weight parameter is shared among multiple layers Signed-off-by:Deepak Narayanan <dnarayanan@nvidia.com>
-
- 28 Nov, 2023 2 commits
-
-
Marks101 authored
* [PyTorch] Linear: fix computation for wgrad if sequence_parallel=True Signed-off-by:
-
Deepak Narayanan authored
Getting warnings of the following form using ToT TE: ``` /usr/local/lib/python3.10/dist-packages/transformer_engine/pytorch/attention.py:852: UserWarning: TypedStorage is deprecated. It will be removed in the future and UntypedStorage will be the only storage class. This should only matter to you if you are using storages directly. To access UntypedStorage directly, use tensor.untyped_storage() instead of tensor.storage() data_ptr = grad_outputs[0].storage().data_ptr() ``` Signed-off-by:Deepak Narayanan <2724038+deepakn94@users.noreply.github.com>
-
- 17 Nov, 2023 2 commits
-
-
cyanguwa authored
* disable FAv2.1 if causal+cross attn Signed-off-by:
-
Kirthi Shankar Sivamani authored
* Delay caching of transposes Signed-off-by:
-
- 15 Nov, 2023 1 commit
-
-
cyanguwa authored
* fix condition checks related to FA head_dim Signed-off-by:
-
- 13 Nov, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
Improve PyTorch memory usage Signed-off-by:
-
- 09 Nov, 2023 1 commit
-
-
Sangkug Lym authored
* Make user buffer name configurable Signed-off-by:
-
- 08 Nov, 2023 1 commit
-
-
Selvaraj Anandaraj authored
* Returning an empty tensor of param dtype for wgrad Signed-off-by:
Selvaraj Anandaraj <selvaraja@computelab-frontend-4-ub22.nvidia.com> * lint Signed-off-by:
-
- 03 Nov, 2023 1 commit
-
-
Xiaowei Ren authored
fix bwd error with FA v2 Signed-off-by:
Xiaowei Ren <xren@nvidia.com> Co-authored-by:
-
- 31 Oct, 2023 1 commit
-
-
Tim Moon authored
* Experimental FP8 tensor Co-authored-by:
Sudhakar Singh <sudhakars@nvidia.com> Co-authored-by:
-
- 24 Oct, 2023 1 commit
-
-
Tim Moon authored
* Do not include logging macros in installed C headers Signed-off-by:
Jan Bielak <jbielak@nvidia.com> Signed-off-by:
-
- 23 Oct, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
* initial test fix Signed-off-by:
-
- 20 Oct, 2023 2 commits
-
-
Tim Moon authored
Signed-off-by:
-
Przemyslaw Tredak authored
* Ability to check cuDNN version from Python Signed-off-by:
-
- 17 Oct, 2023 1 commit
-
-
Kirthi Shankar Sivamani authored
Improve docs Signed-off-by:
-
- 12 Oct, 2023 2 commits
-
-
Kirthi Shankar Sivamani authored
* Add class for RNG state tracker. Signed-off-by:
-
Tim Moon authored
* Debug PyTorch and Paddle tests on Ada Signed-off-by:
-
- 11 Oct, 2023 2 commits
-
-
Xiaowei Ren authored
* rename set_context_parallel_running to set_context_parallel_group Signed-off-by:
-
Kirthi Shankar Sivamani authored
Inference params support Signed-off-by:
-
